Absolute configuration of labdanes and ent-clerodanes from Chromolaena pulchella by vibrational circular dichroism

Phytochemistry 72 (2011) 409–414

Contents lists available at ScienceDirect

Phytochemistry

journal homepage: www.elsevier .com/locate /phytochem

Absolute configuration of labdanes and ent-clerodanes from Chromolaenapulchella by vibrational circular dichroism

Mario A. Gómez-Hurtado a, J. Martín Torres-Valencia a,⇑, Jesús Manríquez-Torres a, Rosa E. del Río b,Virginia Motilva c, Sofía García-Mauriño d, Javier Ávila c, Elena Talero c, Carlos M. Cerda-García-Rojas e,⇑,Pedro Joseph-Nathan e

a Área Académica de Química, Universidad Autónoma del Estado de Hidalgo, Km 4.5 Carretera Pachuca-Tulancingo, Mineral de la Reforma, Hidalgo 42184, Mexicob Instituto de Investigaciones Químico-Biológicas, Universidad Michoacana de San Nicolás de Hidalgo, Apartado 137, Morelia, Michoacán 58000, Mexicoc Facultad de Farmacia, Universidad de Sevilla, Profesor García González No. 2, Sevilla 41012, Spaind Facultad de Biología, Universidad de Sevilla, Profesor García González No. 2, Sevilla 41012, Spaine Departamento de Química, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, Apartado 14-740, México, D.F., 07000 Mexico

a r t i c l e i n f o

Article history:Received 4 November 2010Received in revised form 8 January 2011Accepted 12 January 2011

Keywords:Chromolaena pulchellaAsteraceaeent-ClerodanesLabdanesAbsolute configurationVibrational circular dichroism

0031-9422/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.phytochem.2011.01.021

⇑ Corresponding authors. Tel.: +52 771 717 2002000x6502 (J.M. Torres-Valencia), tel.: +52 55 5747 4(C.M. Cerda-García-Rojas).

E-mail addresses: [email protected] (J.M. Torrtav.mx (C.M. Cerda-García-Rojas).

a b s t r a c t

The aerial parts of Chromolaena pulchella biosynthesize two groups of diterpenes belonging to oppositeenantiomeric series, specifically, the furanoid ent-clerodanes (5R,8R,9S,10R)-(�)-hardwikiic acid (1),methyl (5R,8R,9S,10R)-(�)-hardwikiate (2), (5S,8R,9S,10R)-(�)-hautriwaic acid lactone (3), methyl(5R,8R,9S,10R)-(�)-nidoresedate (4) and methyl (8R,9R)-(�)-strictate (5), as well as the labdanes(5S,8R,9R,10S)-(+)-(13E)-labd-13-ene-8,15-diol (6) and (5S,8R,9R,10S)-(+)-isoabienol (7). The absoluteconfiguration of the two groups of diterpenes was unambiguously assigned by comparison of the vibra-tional circular dichroism spectra of 3 for ent-clerodanes, and of 7 for labdanes with their theoretical spec-tra obtained by density functional theory calculations. The results support a biogenetic proposal toditerpenes found in the studied botanical species.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction natural products, in vitro biomimetic synthesis of a clerodane skel-

The occurrence of natural clerodane and labdane diterpenes hasbeen reported in several plants and liverworts (Merritt and Ley,1992; Blechschmidt and Becker, 1992; Hanson, 2007). Clerodanesand labdanes with related structures are often isolated from thesame organism, since it is considered that the clerodane skeletonbiogenetically derives from a labdane via series of methyl and hy-dride shifts (Dewick, 2009). In some cases, a species can elaborateboth the normal labdane-type and the ent-clerodane-type (neocle-rodane) diterpenoids (Nagashima et al., 1998; Rijo et al., 2002). It isalso of interest that cis and trans ent-clerodane compounds areknown (Merritt and Ley, 1992) and that the stereochemistry ofsome furanoid ent-clerodanes has been revised (Pandey et al.,1984). Hardwickiic acid (1) is a clerodane-type diterpene fromwhich both (�)- and (+)-stereoisomers have been described as nat-ural products (Misra et al., 1979; Costa et al., 1998). Despite thestrong biogenetic link between labdane and clerodane diterpenoid

ll rights reserved.

0x2205; fax: +52 771 717035; fax: +52 55 5747 7137

es-Valencia), ccerda@cinves-

eton from a labdane has not yet been accomplished (George et al.,2010).

Chromolaena pulchella (H.B.K.) R.M. King & H. Rob. (Asteraceae)is an herb found in humid habitats of the Mexican pine oaks forestand plains in the states of Jalisco, Colima, Michoacán, Guerrero,Puebla and México. No reports regarding the traditional use for thisspecies have been published, although etnobotanical studies docu-mented by us in the state of Michoacán established this plant hasbeen employed as a traditional remedy for gastrointestinal disor-ders and as an antipyretic and anti-inflammatory agent. In thiswork, it was found that C. pulchella mainly biosynthesizes ent-clerodanes but also labdanes belonging to the opposite enantio-meric series, thus representing an interesting case for biogeneticstudies. The absolute configuration of the two diterpene series isverified by vibrational circular dichroism spectroscopy in thisstudy.

2. Results and discussion

Systematic column chromatography fractionation of hexaneand CH2Cl2 extracts from the aerial parts of C. pulchella lead tothe isolation of the furanoid ent-clerodanes (�)-hardwikiic acid

Scheme 1. Proposed biosynthetic pathway for ent-clerodanes 1–5 and labdanes 6and 7 from Chromolaena pulchella.

410 M.A. Gómez-Hurtado et al. / Phytochemistry 72 (2011) 409–414

(1) (Misra et al., 1979), methyl (�)-hardwikiate (2) (Misra et al.,1979), (�)-hautriwaic acid lactone (3) (Bohlmann and Grenz,1972; Hsü et al., 1971), methyl (�)-nidoresedate (4) (Pandeyet al., 1984), methyl (�)-strictate (5) (Tandon and Rastogi, 1979;Pandey et al., 1984), together with (+)-(13E)-labd-13-ene-8,15-diol(6) (Nagashima et al., 1998; Forster et al., 1985), and (+)-isoabienol(7) (Nagashima et al., 1998; Cabrera et al., 1983). In addition, b-caryophyllene oxide (Krebs et al., 1990), sphatulenol (Tinto et al.,1992), eudesm-4(15)-en-1b,6a-diol (Sun et al., 2004), lupeoyl stea-rate (Razdan et al., 1996), a-amyrin and b-amyrin stearates (Men-des et al., 1999), b-sitosterol 3-O-b-D-glucoside (Ahmad et al.,2003), stigmasterol 3-O-b-D-glucoside (Ahmad et al., 2003), b-sitosterol, stigmasterol and stigmastenol were also isolated fromthis source.

Identification of all components was supported by spectro-scopic analyses including 1D and 2D NMR spectroscopy in compar-ison with published data. Methyl (�)-hardwickiate (2) wasidentified as a minor compound in fractions containing a mixtureof this diterpene and methyl nidoresedate (4). In addition, com-pound 2 was prepared from (�)-hardwickiic acid (1) using diazo-methane, whose spectroscopic data were equal to those reported(Misra et al., 1979). This is the first time that diterpenes 1–7 areisolated from the genus Chromolaena.

The closely related structures of ent-clerodanes 1–5 and labd-anes 6 and 7 can be integrated into the biosynthetic relationshipoutlined in Scheme 1. Geranylgeranyl pyrophosphate (GGPP) un-der protonation initiates two enantiodivergent concerted cycliza-tion sequences to lead either labdanyl or ent-labdanyl cations.The labdanyl cation is quenched by addition of water to give(13E)-labd-13-ene-8,15-diol (6), which can provide isoabienol (7)by water loss. On the other hand, ent-labdanyl cation may precip-itate a series of concerted Wagner–Meerwein hydride and methylmigrations, resulting in the rearranged ent-clerodanyl cation,which may lead to hardwickiic acid (1) by proton loss from C-3 fol-lowed by a series of oxidative steps to generate the carboxyl groupat C-19 and the furan ring in the side-chain. Hardwickiic acid (1)can undergo methylation to afford the corresponding methyl ester2, or can experience oxidation at C-18 to form the intermediatehautriwaic acid (Hsü et al., 1971), which after lactonization gives3. Alternatively, allylic hydroxylation at C-2 may provide interme-diate 2-hydroxyhardwickiic acid (Pandey et al., 1984), which inturn can generate nidoresedic acid by water loss. Methyl ester for-mation from the last can give methyl nidoresedate (4), whose elec-trocyclic ring opening reaction of the 1,3-cyclohexadiene moiety(Bohlmann et al., 1983; Pandey et al., 1984) leads to methyl stric-tate (5).

In order to support the proposed biogenetic scheme, whichpoints out that in C. pulchella labdanes 6 and 7 are not precursorsof furoclerodanes 1–5, it was crucial to secure the absolute config-uration of the two groups of diterpenes. This task can be accom-plished if a selected representative of each type of compounds isstudied by vibrational circular dichroism (VCD) spectroscopy(Freedman et al., 2003). This technique, which has been recentlyused to determine the absolute configuration of natural products(Nafie, 2008; Cerda-García-Rojas et al., 2008; Reina et al., 2010;Muñoz et al., 2010), is based on the comparison between theexperimental VCD spectrum and the corresponding theoreticalcurve, for the proper enantiomer, obtained by density functionaltheory (DFT) calculations. The calculations involve generation ofweighted-averaged vibrational plots including all significantlypopulated conformations of the analyzed molecule. Thus, ent-clerodane 3 and a labdane 7 were used as reference compoundsto provide certainty to the absolute configuration of each groupof diterpenes using the VCD calculations.

The molecular models of 3 and 7 were built according to theabsolute configuration proposed in Scheme 1 and subjected to a

conformational search using the Monte Carlo method and theMMFF molecular mechanics force field. The searching processwas repeated several times starting from different geometries inorder to cover all the conformational hypersurface. The obtainedconformers were analyzed and filtered according to their energies,removing all duplicates. A selection between 0 and 10 kcal/molafforded 25 conformations for 3 and 57 conformations for 7. DFTenergy calculation of the selection within a 0–5 kcal/mol range atthe B3LYP/6-31G(d) level of theory left 12 conformers for com-pound 3 and 29 conformers for compound 7. Geometry optimiza-tion using the B3LYP/DGDZVP level of theory and calculation offrequencies, after narrowing the energy range 0–2 kcal/mol,yielded only two conformers for hautriwaic acid lactone (3) andnine conformers for isoabienol (7).

Table 1 contains the thermochemical analysis for the calcula-tion of the conformational equilibrium of 3 which included thetwo main species 3a and 3b and several minor conformers 3c–3f.The Boltzmann distribution was calculated according to theDG = DH–TDS and DG = �RT ln K equations, considering theB3LYP/DGDZVP calculated frequencies. The two low-energy struc-tures, accounting for 96.5% of the conformational population, arerepresented in Fig. 1. Both species essentially preserve the samespatial arrangements with changes in rotation of the C12–C13 bond.

Table 1Thermochemical analysis of (5S,8R,9S,10R)-(�)-hautriwaic acid lactone (3).

Conformer DEMMFFa %MMFF

b DEDFTc %DFT

d DGOPTe %OPT

f

3a 0.00 49.0 0.00 50.4 0.00 56.13b 0.01 48.2 0.06 45.3 0.20 40.43c 3.84 0.1 2.44 0.8 2.27 1.23d 2.21 1.2 2.19 1.2 2.43g 0.93e 3.85 0.1 2.42 0.8 2.44 0.93f 2.08 1.4 2.13 1.4 2.85g 0.5

a Molecular mechanics energies relative to 3a with EMMFF = 70.129 kcal/mol.b Population in % calculated from the MMFF energies according to DEMMFF � �RT

ln K.c Single point B3LYP/6-31G(d) energies relative to 3a with E6–

31G(d) = �629651.512 kcal/mol.d Population in % calculated from B3LYP/6-31G(d) energies according to DE6–

31G(d) � �RT ln K.e Gibbs free energies relative to 3a with GDGDZVP = �629484.989 kcal/mol.f Population in % calculated from Gibbs free energies according to DG = �RT ln K.g Estimated by taking into account the averaged entropic term of conformers 3a–

c and 3e.

-150

-120

-90

-60

-30

0

30

9501050115012501350145015501650

-40

-20

0

20

40

60

Δε10

3

Wavenumber (cm-1)

(a)

(b)

Fig. 2. (a) Experimental and (b) B3LYP/DGDZVP DFT Boltzmann-weighted VCDspectra of (5S,8R,9S,10R)-(�)-hautriwaic acid lactone (3).

Mol

ar a

bsor

ptiv

ityε

(a)

(b)

0.0

0.1

0.0

0.1

0.2

0.3

0.4

M.A. Gómez-Hurtado et al. / Phytochemistry 72 (2011) 409–414 411

The VCD frequencies of the two major conformers 3a and 3b werecalculated and Boltzmann-averaged to generate the calculatedspectrum which was plotted using Lorentzian bandshapes andbandwidths of 6 cm�1. Fig. 2 shows the comparison between thetheoretical and experimental VCD spectra of hautriwaic acid lac-tone (3), which showed a very good agreement. Quantitative eval-uation of this concordance was achieved by applying a recentlydeveloped confidence level algorithm (BioTools Co., Jupiter, FL33458, USA), which calculates the integrated overlap of the exper-imental and theoretical data as a function of a relative shift. Appli-cation of this procedure allowed us to obtain the optimalanharmonicity factor (anH = 0.980), and the VCD spectral similarityfor the correct (SE = 80.8%) and the incorrect enantiomer(S�E = 9.1%) with a 100% confidence level for the absolute configu-ration determination. Further details of the algorithm applicationhave been reported in the VCD study of (�)-myrtenal (Burgueño-Tapia et al., 2010). In addition, Fig. 3 shows the comparison be-tween the experimental and calculated IR spectra of 3.

For the determination of the absolute configuration of (+)-isoa-bienol (7), the same procedure was applied. In this case, a higherconformational flexibility was observed. Table 2 contains the ther-mochemical analysis for the calculation of the conformationalequilibrium of 7 composed by the nine more relevant species7a–7i and several less contributing conformers 7j–7n. The ninemost stable structures, accounting for 98.6% of the conformationalpopulation, are illustrated in Fig. 4. These conformers conservedthe same spatial arrangement for the bicyclic skeleton but showedsignificant variations in the hexadienyl side chain at C9 togetherwith rotation of the C8–O8 bond. The VCD frequencies of the se-

Fig. 1. Main DFT B3LYP/DGDZVP minimum energy structures of (5S,8R,9S,10R)-(�)-hautriwaic acid lactone (3).

Wavenumber (cm-1)

0.2

0.3

0.49501050115012501350145015501650

Fig. 3. (a) Experimental and (b) B3LYP/DGDZVP DFT Boltzmann-weighted IRspectra of (5S,8R,9S,10R)-(�)-hautriwaic acid lactone (3).

lected conformers (7a and 7i) were Boltzmann-averaged to gener-ate the corresponding calculated spectrum shown in Fig. 5 togetherwith the experimental spectrum. In this case, application of theconfidence level algorithm afforded the optimal anharmonicityfactor (anH = 0.969), and the VCD spectral similarity for the correct

Table 2Thermochemical analysis of (5S,8R,9R,10S)-(+)-isoabienol (7).

Conformer DEMMFFa %MMFF

b DEDFTc %DFT

d DGOPTe %OPT

f

7a 1.55 2.5 0.60 12.6 0.00 27.97b 0.00 33.5 1.64 2.2 0.16 21.47c 0.25 22.0 1.14 5.0 0.29 17.17d 0.22 22.9 0.98 6.6 0.62 9.77e 1.39 3.2 1.33 3.6 0.75 7.87f 2.05 1.1 0.00 34.4 0.79 7.47g 1.91 1.3 0.34 19.4 1.08 4.57h 1.05 5.7 1.28 4.0 1.61 1.87i 1.63 2.1 3.56 0.1 1.98 1.07j 3.15 0.2 2.73 0.3 2.19 0.77k 2.55 0.5 0.97 6.7 2.69 0.37l 1.88 1.4 1.89 1.4 3.00 0.27m 2.95 0.2 1.77 1.7 3.13 0.17n 2.87 0.3 2.00 1.2 3.25 0.1

a Molecular mechanics energies relative to 7b with EMMFF = 80.030 kcal/mol.b Population in % calculated from the MMFF energies according to DEMMFF � �RT

ln K.c Single point B3LYP/6-31G(d) energies relative to 7f with E6–

31G(d) = �538269.212 kcal/mol.d Population in % calculated from B3LYP/6-31G(d) energies according to DE6–

31G(d) � �RT ln K.e Gibbs free energies relative to 7a with GDGDZVP = �538036.161 kcal/mol.f Population in % calculated from Gibbs free energies according to DG = �RT ln K.

412 M.A. Gómez-Hurtado et al. / Phytochemistry 72 (2011) 409–414

(SE = 77.5%) and the incorrect enantiomer (S�E = 25.3%) with a 100%confidence level for the absolute configuration determination. Fi-nally, Fig. 6 shows the comparison between the experimentaland calculated IR spectra of (+)-isoabienol (7).

Fig. 4. Main DFT B3LYP/DGDZVP minimum energy

3. Conclusion

The results presented herein indicate that the experimentalVCD curves of (�)-hautriwaic acid lactone (3) and (+)-isoabienol(7) obtained from C. pulchella show good agreement with the DFTB3LYP/DGDZVP VCD calculated spectra, as evidenced in Figs. 2and 5, respectively, thus affording conclusive and direct evidencefor the absolute configuration of these compounds as follows:(5R,8R,9S,10R)-(�)-hardwikiic acid (1), methyl (5R,8R,9S,10R)-(�)-hardwikiate (2), (5S,8R,9S,10R)-(�)-hautriwaic acid lactone (3),methyl (5R,8R,9S,10R)-(�)-nidoresedate (4), methyl (8R,9R)-(�)-strictate (5), (5S,8R,9R,10S)-(+)-(13E)-labd-13-ene-8,15-diol (6)and (5S,8R,9R,10S)-(+)-isoabienol (7). This theoretical and experi-mental protocol leads to the conclusion that C. pulchella biosynthe-sizes both the normal labdane-type and the ent-clerodane-typediterpenoids supporting the biogenetic pathway depicted inScheme 1.

4. Experimental

4.1. General experimental procedures

Melting points were determined on a Fisher–Johns apparatusand are uncorrected. Optical rotations were determined in CHCl3

on a Perkin–Elmer 341 polarimeter. UV spectra were determinedon a Perkin–Elmer Lambda 12 UV/vis spectrophotometer. IR spec-tra were recorded on a Perkin–Elmer 16F PC IR-FT spectrophotom-eter using thin films of the compounds deposited on a CsI crystal.

structures of (5S,8R,9R,10S)-(+)-isoabienol (7).

Δε10

3

Wavenumbers (cm-1)

-40

-20

0

20

40

-40

-20

0

20

40

95010501150125013501450155016501750

(b)

(a)

Fig. 5. (a) Experimental and (b) B3LYP/DGDZVP DFT Boltzmann-weighted VCDspectra of (5S,8R,9R,10S)-(+)-isoabienol (7).

Wavenumber (cm-1)

Mol

ar a

bsor

ptiv

ityε

(a)

(b)

0.0

0.1

0.2

0.3

0.4

0.5

0.0

0.1

0.2

0.3

0.4

0.595010501150125013501450155016501750

Fig. 6. (a) Experimental and (b) B3LYP/DGDZVP DFT Boltzmann-weighted IRspectra of (5S,8R,9R,10S)-(+)-isoabienol (7).

M.A. Gómez-Hurtado et al. / Phytochemistry 72 (2011) 409–414 413

VCD and IR measurements for 3 and 7 were performed on a Bio-Tools ChiralIR FT-VCD spectrophotometer equipped with dualphotoelastic modulation. Samples of 3 and 7 (9 and 14 mg, respec-tively) were dissolved in CDCl3 (150 lL) and placed in a BaF2 cellwith a path length of 100 lm. Data were acquired at a resolution

of 4 cm�1 during 4 h for each compound. NMR measurements,including COSY, HMQC, and HMBC experiments, were performedat 400 MHz for 1H and 100 MHz for 13C on a JEOL Eclipse 400 spec-trometer from CDCl3 solutions using TMS as internal standard.Mass spectra were recorded at 70 eV on a Hewlett–Packard 5890Series II spectrometer and on a Hewlett–Packard 5989A spectrom-eter. Column chromatography was carried out on Merck silica gel60 (Aldrich, 230–400 mesh ASTM).

4.2. Molecular modeling

Geometry optimizations for 3 and 7 were carried out using theMMFF94 force-field calculations as implemented in the Spartan’04program. A Monte Carlo search protocol was carried out consider-ing an energy cutoff of 10 kcal/mol. A systematic search was alsoperformed, but no additional minimum energy structures werefound. Single point energy of the conformers found within theEMMFF 0–10 kcal/mol range was calculated with the DFT B3LYP/6-31G(d) level of theory in the Spartan’04 program and DFT conform-ers found in the first 5 kcal/mol were optimized at the B3LYP/DGDZVP level of theory employing the Gaussian 03 W program.The minimized structures in the first 3 kcal/mol were used to cal-culate the thermochemical parameters and the IR and VCD fre-quencies at 298 K and 1 atm. The VCD spectra were Boltzmann-averaged using those conformers within 0–2 kcal/mol. Molecularvisualization was carried out with the GaussianView 3.0 program.Calculations required between 17 and 24 h computational time perconformer when using a desktop computer operating at 3 GHzwith 8 Gb RAM.

4.3. Plant material

Specimens of C. pulchella were collected near km 61 of Morelia-Zacapu state road No. 15, in the municipality of Constitución, Stateof Michoacán, Mexico, during October 2005. A voucher specimen(No. 192522) is deposited at the Herbarium of the Instituto de Eco-logía, A.C., Centro Regional del Bajío, Pátzcuaro, Michoacán, Méxi-co, where Prof. Jerzy Rzedowski kindly identified the plantmaterial.

4.4. Extraction and isolation

Air-dried flowers of C. pulchella (280 g) were extracted withCH2Cl2 (2 L) under conditions of reflux for 6 h. Filtration and evap-oration of the extract afforded a yellow viscous oil (29 g). A portion(1 g) was applied to silica gel (120 g) using hexane–CHCl3 (1:1),CHCl3, CHCl3–acetone (4:1), CHCl3–acetone (7:3), and acetone aseluents. Fractions of 100 mL of each polarity were collected, mon-itored by TLC and analyzed by 1H NMR spectroscopy. The resultingmaterial from each fraction was marked as A (105 mg), B (120 mg),C (125 mg), D (190 mg), and E (55 mg). This procedure was re-peated five times. Fatty materials were identified from fraction A,fraction C gave a mixture of b-sitosterol and stigmasterol, whilefraction D afforded a mixture of b-sitosterol 3-O-b-D-glucosideand stigmasterol 3-O-b-D-glucoside. Minor and inseparable organicmaterials were observed in fraction E. Fraction B (150 mg) wassubjected to silica gel CC (5 g) using a hexane–EtOAc gradient(49:1, 24:1, 12:1 and 9:1, v/v) as eluent. Elution with hexane–EtOAc (49:1) afforded 5 (4 mg), [a]D

20 �79(c 0.5, CHCl3), elutionwith hexane–EtOAc (24:1) gave 7 (19.5 mg), [a]D

20 + 4 (c 1.0,CHCl3), elution with hexane–EtOAc (23:2) yielded 1 (45 mg),[a]D

20 �87 (c 0.8, CHCl3), and 4 (12.5 mg) [a]D20 �147 (c 1.2,

CHCl3), which contained 17% of 2. Pure 2, prepared by treatmentof 1 with diazomethane, showed [a]D

20 �81 (c 0.8, CHCl3). Finally,elution with hexane–EtOAc (9:1) provided 3 (5 mg) of [a]D

20 �77(c 0.4, CHCl3).

414 M.A. Gómez-Hurtado et al. / Phytochemistry 72 (2011) 409–414

Air-dried and powdered leaves of C. pulchella (425 g) were ex-tracted with hexane (3.5 L) under conditions of reflux for 6 h. Fil-tration and evaporation of the extract afforded a greenish,viscous oil (14.7 g), which was dissolved in MeOH at 50 �C, thenkept at 0 �C for 12 h, and filtered to remove fatty materials. The fil-tered solution was evaporated in vacuo and a portion of the residue(1 g) was applied to a silica gel column (120 g) using hexane, hex-ane–EtOAc (6:1), hexane–EtOAc (7:3), and EtOAc as eluents. Frac-tions of 250 mL of each polarity were collected and marked as A(160 mg), B (61 mg), C (5 mg), and D (50 mg). This procedure wasrepeated five times. TLC and 1H NMR spectroscopic analysis ofthese fractions showed the major constituents were: fraction A,b-caryophyllene oxide, a-amyrin stearate, and b-amyrin stearate.Fraction B gave 7, eudesm-4(15)-en-1b,6a-diol, spathulenol, stig-masterol, and stigmastenol. Fraction C afforded 6 (5 mg),[a]D

20 + 1.8 (c 0.9, CHCl3). Fatty materials were observed in fractionD. Fraction A (150 mg) was rechromatographed over silica gel (5 g).Elution with hexane–EtOAc (49:1) afforded pure b-caryophylleneoxide (9.5 mg). Alternatively, fraction A (250 mg) was purified ona silica gel–AgNO3 (6:1) (8.5 g) CC. Elution with hexane–CHCl3

(47:3) afforded a mixture of a-amyrin and b-amyrin estearates(35 mg), whereas elution with CHCl3 gave lupeoyl stearate(90 mg). Fraction B (141 mg) was re-applied to a silica gel column(5 g) using CHCl3 and CHCl3–acetone (24:1) as eluents. Elutionwith CHCl3 gave 7 (11 mg), while elution with CHCl3–acetoneyielded eudesm-4(15)-en-1b,6a-diol (6.5 mg) and spathulenol(11.8 mg). Finally, fraction B (180 mg) was purified on a silicagel–AgNO3 (6:1) (8.5 g) CC eluting with hexane–AcOEt (9:1) to af-ford stigmasterol (3 mg) and stigmastenol (3.5 mg).

Acknowledgements

Partial financial support from CONACYT, Mexico (Grants No.U2-80555 and 11828), and from CIC-UMSNH is acknowledged.MAGH and JMT thank CONACYT for fellowships 206545 and171415, respectively. We are grateful to Professor Jerzy Rzedowski(Instituto de Ecología, A.C., Centro Regional del Bajío, Pátzcuaro,Michoacán, Mexico) for identifying the plant material.

References

Ahmad, B., Jan, Q., Bashir, S., Choudhary, M.I., Nisar, M., 2003. Phytochemicalevaluation of Chenopodium murale Linn. Asian J. Plant Sci. 2, 1072–1078.

Blechschmidt, M., Becker, H., 1992. Ent-labdanes and furanoditerpenes from theliverwort Jamesoniella autumnalis. J. Nat. Prod. 55, 111–121.

Bohlmann, F., Grenz, M., 1972. Notiz über die Isolierung von Hautriwa-Säure ausConyza ivaefolia Less. Chem. Ber. 105, 3123–3125.

Bohlmann, F., Grenz, M., Wegner, P., Jakupovic, J., 1983. Clerodane derivatives andnovel diterpenes from Conyza scabrida DC. Liebigs Ann. Chem. 11, 2008–2020.

Burgueño-Tapia, E., Zepeda, L.G., Joseph-Nathan, P., 2010. Absolute configuration of(�)-myrtenal by vibrational circular dichroism. Phytochemistry 71, 1158–1161.

Cabrera, E., García-Granados, A., Sáenz-de-Buruaga, A., Sáenz-de-Buruaga, J.M.,1983. Diterpenoids from Sideritis hirsuta subsp. Nivalis. Phytochemistry 22,2779–2781.

Cerda-García-Rojas, C.M., Catalán, C.A.N., Muro, A.C., Joseph-Nathan, P., 2008.Vibrational circular dichroism of africanane and lippifoliane sesquiterpenesfrom Lippia integrifolia. J. Nat. Prod. 71, 967–971.

Costa, M., Fujiwara, F.Y., Imamura, P.M., 1998. Assignment of 13C NMR data ofmethyl (+)-hardwickiate and its derivatives. Magn. Reson. Chem. 36, 542–544.

Dewick, P.M., 2009. Medicinal Natural Products. A Biosynthetic Approach, 3rd ed.John Wiley & Sons, Chichester. pp. 233–234.

Forster, P.G., Ghisalberti, E.L., Jefferies, P.R., 1985. Labdane diterpenes from an Acaciaspecies. Phytochemistry 24, 2991–2993.

Freedman, T.B., Cao, X., Dukor, R.K., Nafie, L.A., 2003. Absolute configurationdetermination of chiral molecules in the solution state using vibrational circulardichroism. Chirality 15, 743–758.

George, J.H., McArdle, M., Baldwin, J.E., Adlington, R.M., 2010. Biomimeticrearrangements of simplified diterpenoids. Tetrahedron 66, 6321–6330.

Hanson, J.R., 2007. Diterpenoids. Nat. Prod. Rep. 24, 1332–1341. Earlier reviews:2006, 23, 875–885; 2005, 22, 594–602; 2004, 21, 785–793; 2004, 21, 312–320;2003, 20, 70–78; 2002, 19, 125–132; 2001, 18, 88–94; 2000, 17, 165–174.

Hsü, H.-Y., Chen, Y.P., Kakisawa, H., 1971. Structure of hautriwaic acid.Phytochemistry 10, 2813–2814.

Krebs, H.C., Rakotoarimanga, J.V., Habermehl, G.G., 1990. Isolation of spathulenoland (�)-caryophyllene oxide from Vernonia mollissima Don and 1H and 13Creassignment by two-dimensional NMR spectroscopy. Magn. Reson. Chem. 28,124–128.

Mendes, C.C., Cruz, F.G., David, J.M., Nascimento, I.P., David, J.P., 1999. Triterpenosestereficados com ácidos graxos e ácidos triterpénicos isolados de Byrsonimamicrophylla. Quím. Nova 22, 185–188.

Merritt, A.T., Ley, S.V., 1992. Clerodane diterpenoids. Nat. Prod. Rep. 9, 243–287.Misra, R., Pandey, R.C., Dev, S., 1979. Higher isoprenoids-VIII. Diterpenoids from the

oleoresin of Hardwickia pinnata part 1: hardwickiic acid. Tetrahedron 35, 2301–2310.

Muñoz, M.A., Areche, C., Rovirosa, J., San-Martin, A., Joseph-Nathan, P., 2010.Absolute configuration of sargaol acetate using DFT calculations and vibrationalcircular dichroism. Heterocycles 81, 625–635.

Nafie, L.A., 2008. Vibrational circular dichroism: a new tool for the solution-statedetermination of the structure and absolute configuration of chiral naturalproducts molecules. Nat. Prod. Commun. 3, 451–466.

Nagashima, F., Takaoka, S., Asakawa, Y., 1998. Diterpenoids from the Japaneseliverwort Jungermannia infusca. Phytochemistry 49, 601–608.

Pandey, U.C., Singhal, A.K., Barua, N.C., Sharma, R.P., Baruah, J.N., Watanabe, B.K.,Kulanthaivel, P., Herz, W., 1984. Stereochemistry of strictic acid and relatedfurano-diterpenes from Conyza japonica and Grangea maderaspatana.Phytochemistry 23, 391–397.

Razdan, T.K., Kachroo, P.K., Qurishi, M.A., Kalla, A.K., Waight, E.S., 1996. Unusualhomologous long-chain alkanoic acid esters of lupeol from Koelpinia linearis.Phytochemistry 41, 1437–1438.

Reina, M., Burgueño-Tapia, E., Bucio, M.A., Joseph-Nathan, P., 2010. Absoluteconfiguration of tropane alkaloids from Schizanthus species by vibrationalcircular dichroism. Phytochemistry 71, 810–815.

Rijo, P., Gaspar-Manríquez, C., Fátima-Simões, M., Duarte, A., Aprenda-Rojas, M.C.,Cano, F.H., Rodríguez, B., 2002. Neoclerodane and labdane diterpenoids fromPlectranthus ornatus. J. Nat. Prod. 65, 1387–1390.

Sun, Z., Chen, B., Zhang, S., Hu, C., 2004. Four new eudesmanes from Caraganaintermedia and their biological activities. J. Nat. Prod. 67, 1975–1979.

Tandon, S., Rastogi, R.P., 1979. Strictic acid, a novel diterpene from Conyza stricta.Phytochemistry 18, 494–495.

Tinto, W.F., Blair, L.C., Reynolds, W.F., McLean, S., 1992. Terpenoid constituents ofOxandra asbeckii. J. Nat. Prod. 55, 701–706.